slide r.6 - 1 copyright © 2008 pearson education, inc. publishing as pearson addison-wesley
TRANSCRIPT
Slide R.6 - 1 Copyright © 2008 Pearson Education, Inc. Publishing as Pearson Addison-Wesley
Copyright © 2008 Pearson Education, Inc. Publishing as Pearson Addison-Wesley
Mathematical Models and Curve Fitting
OBJECTIVE Use curve fitting to find a mathematical
model for a set of data and use the model to make predictions.
R.6
Slide R.6 - 3 Copyright © 2008 Pearson Education, Inc. Publishing as Pearson Addison-Wesley
Example 1: For the scatterplots and graphs on the following slides, determine which, if any, of the following functions might be used as a model for the data.
Linear, f (x) = mx + bQuadratic, f (x) = ax2 + bx + c, a > 0Quadratic, f (x) = ax2 + bx + c, a < 0Polynomial, neither linear nor quadratic
R.6 Mathematical Models and Curve Fitting
Slide R.6 - 4 Copyright © 2008 Pearson Education, Inc. Publishing as Pearson Addison-Wesley
Example 1 (continued): a)
The data rise and then fall in a curved manner, fitting a quadratic function,
f (x) = ax2 + bx + c, a < 0.
R.6 Mathematical Models and Curve Fitting
Slide R.6 - 5 Copyright © 2008 Pearson Education, Inc. Publishing as Pearson Addison-Wesley
Example 1 (continued): b)
The data seem to fit a linear function,f (x) = mx + b.
R.6 Mathematical Models and Curve Fitting
Slide R.6 - 6 Copyright © 2008 Pearson Education, Inc. Publishing as Pearson Addison-Wesley
Example 1 (continued): c)
The data rise in a manner fitting the right-hand side of a quadratic function,
f (x) = ax2 + bx + c, a > 0.
R.6 Mathematical Models and Curve Fitting
Slide R.6 - 7 Copyright © 2008 Pearson Education, Inc. Publishing as Pearson Addison-Wesley
Example 1 (continued): d)
The data fall and then rise in a curved manner, fitting a quadratic function,
f (x) = ax2 + bx + c, a > 0.
R.6 Mathematical Models and Curve Fitting
Slide R.6 - 8 Copyright © 2008 Pearson Education, Inc. Publishing as Pearson Addison-Wesley
Example 1 (continued): e)The data rise and fall more than once, so they do not fit a linear or quadratic function but might fit a polynomial function that is neither quadratic nor linear.
R.6 Mathematical Models and Curve Fitting
Slide R.6 - 9 Copyright © 2008 Pearson Education, Inc. Publishing as Pearson Addison-Wesley
Example 2: The following table shows the annual percent increases in pay since 1996 for a U.S. production worker.
R.6 Mathematical Models and Curve Fitting
Number of years, x,
since 19961 2 3 4 5 6 7
Percent increase
since 1996, P 1.9 7.4 11.7 19.5 28.2 29.7 31.3
Slide R.6 - 10 Copyright © 2008 Pearson Education, Inc. Publishing as Pearson Addison-Wesley
Example 2 (continued):a) Make a scatterplot of the data and determine
whether the data seem to fit a linear function.b) Find a linear function that (approximately) fits the
data.c) Use the model to predict the percentage by which
2010 wages will exceed 1996 wages.
R.6 Mathematical Models and Curve Fitting
Slide R.6 - 11 Copyright © 2008 Pearson Education, Inc. Publishing as Pearson Addison-Wesley
Example 2 (continued):a) The scatterplot is shown
below. The data tend to follow a straight line, although a “perfect” straight line cannot be drawn through the data points.
R.6 Mathematical Models and Curve Fitting
Slide R.6 - 12 Copyright © 2008 Pearson Education, Inc. Publishing as Pearson Addison-Wesley
Example 2 (continued):b) We consider the function P(x) = mx + b, where P(x)
is the percentage by which the wages x years after 1996 exceed the wages in 1996. To derive the constants (or parameters) m and b, we choose two data points. Although this procedure is somewhat arbitrary, we try to choose two points that follow the general linear pattern. In this case, we pick (1, 1.9) and (4, 19.5).
R.6 Mathematical Models and Curve Fitting
Slide R.6 - 13 Copyright © 2008 Pearson Education, Inc. Publishing as Pearson Addison-Wesley
Example 2 (continued):Since the points are to be solutions of the linear equation, it follows that
We now have a system of equations. We solve by subtracting each side of the top equation from each side of the bottom equation to eliminate b.
R.6 Mathematical Models and Curve Fitting
.45.19
1.9or
45.19
19.1
bm
bm
bm
bm
m
m
87.5
36.17
Slide R.6 - 14 Copyright © 2008 Pearson Education, Inc. Publishing as Pearson Addison-Wesley
Example 2 (continued):Substituting 5.87 for m into either equation, we can solve for b.
Thus, we get the equation (model)P(x) = 5.87x – 3.97.
R.6 Mathematical Models and Curve Fitting
b
b
97.3
87.59.1
Slide R.6 - 15 Copyright © 2008 Pearson Education, Inc. Publishing as Pearson Addison-Wesley
Example 2 (concluded):c) The percentage by which U.S. production workers’
wages will have increased in 2010, a total of 14 years after 1996, is
P(14) = 5.87 · 14 – 3.97 = 78.21%.
R.6 Mathematical Models and Curve Fitting
Slide R.6 - 16 Copyright © 2008 Pearson Education, Inc. Publishing as Pearson Addison-Wesley
Example 3: In a study by Dr. Harold J. Morowitz of Yale University, data were gathered that showed the relationship between the death rate of men and the average number of hours per day that the men slept. These data are listed in the table on the next slide.
a) Make a scatterplot of the data, and determine whether the data seem to fit a quadratic function.
b) Find a quadratic function that fits the data.c) Use the model to find the death rate for males who sleep 2
hr, 8 hr, and 10 hr.
R.6 Mathematical Models and Curve Fitting
Slide R.6 - 17 Copyright © 2008 Pearson Education, Inc. Publishing as Pearson Addison-Wesley
Example 3 (continued):
R.6 Mathematical Models and Curve Fitting
Average Number of Hours of Sleep, x
Death Rate per 100,000 Males, y
5 1121
6 805
7 626
8 813
9 967
Slide R.6 - 18 Copyright © 2008 Pearson Education, Inc. Publishing as Pearson Addison-Wesley
Example 3 (continued):a) The scatterplot is shown below. Note that the rate drops
and then rises, which suggests that a quadratic function might fit the data.
R.6 Mathematical Models and Curve Fitting
Slide R.6 - 19 Copyright © 2008 Pearson Education, Inc. Publishing as Pearson Addison-Wesley
Example 3 (continued):b) We consider the quadratic model,
y = ax2 + bx + c.To derive the constants (or parameters) a, b, and c, we use the three data points (5, 1121), (7, 626), and (9, 967). Since these points are to be solutions of a quadratic equation, it follows that
R.6 Mathematical Models and Curve Fitting
.981967
749626
5251121
or
99967
77626
551121
2
2
2
cba
cba
cba
cba
cba
cba
Slide R.6 - 20 Copyright © 2008 Pearson Education, Inc. Publishing as Pearson Addison-Wesley
Example 3 (continued):We solve this system of three equations in three variables using procedures of algebra and get
a = 104.5, b = –1501.5, and c = 601.6,Thus, the quadratic model for this data is given by
y = 104.5x2 – 1501.5x + 601.6.
c) The death rate for males who sleep 2 hr is given byy = 104.5 · 22 – 1501.5 · 2 + 601.6 = 3431.
R.6 Mathematical Models and Curve Fitting
Slide R.6 - 21 Copyright © 2008 Pearson Education, Inc. Publishing as Pearson Addison-Wesley
Example 3 (concluded):The death rate for males who sleep 8 hr is given by
y = 104.5 · 82 – 1501.5 · 8 + 601.6 = 692.
The death rate for males who sleep 10 hr is given byy = 104.5 · 102 – 1501.5 · 10 + 601.6 = 1451.
R.6 Mathematical Models and Curve Fitting